Chemical Composition and Mass Emission Factors of Candle Smoke Particles

Overview

Pagels and colleagues measured candle-smoke particle composition from two tapered candle types during steady burn, sooting burn, and smouldering in a controlled chamber. The study is routeable as home-air emission product context because Table 3 reports metal component emission factors for Cu, tin (Sn), Co, Zn, and lead (Pb) from the candle-smoke particles. The source reports emission factors in µg h-1 and particle-mass emission factors in mg h-1; no product-content concentration, migration value, or consumer threshold is inferred here.

Key numbers

• Sample frame: two tapered candles made by market-leading companies in Sweden. Candle I was a white pure-stearin candle marketed as low-soot and in a higher price range. Candle II was a dark-blue budget candle made from a stearin/paraffin wax mixture.
• Experimental setup: candle smoke was generated in a 21.6 m3 stainless-steel chamber at 30 ± 5% RH and 23-28 °C, with a supply-air flow rate of 10.8 m3 h-1 corresponding to 0.5 h-1 air exchange.
• Burn modes: steady burn with convective-air shields, sooting burn without shields, and smouldering upon extinction. Steady and sooting burn were performed for both candle types; smouldering was performed only for Candle II. All experiments were repeated two to three times.
• Table 1 total PM2.5 mass emission factors for a single candle: Candle I steady burn 2.4 ± 0.1 mg h-1; Candle II steady burn 0.87 ± 0.14 mg h-1; Candle I sooting 8.9 ± 0.4 mg h-1; Candle II sooting 25.3 ± 0.02 mg h-1.
• Table 1 inorganic-compound emission factors: Candle I steady burn 2.9 ± 0.5 mg h-1; Candle II steady burn 0.92 ± 0.34 mg h-1; Candle I sooting 3.3 ± 0.6 mg h-1; Candle II sooting 1.1 ± 0.6 mg h-1.
• Table 2 smouldering emission for Candle II, reported per single extinction of a single candle: PM2.5 mass 0.72 ± 0.04 mg, elemental carbon 0.03 ± 0.01 mg, organic matter 0.80 ± 0.03 mg, and inorganic compounds 0.11± 0.06 mg.
• Figure 5 note: the chamber EC concentration during Candle II sooting burn extended off-scale; the value is reported as 1424 µg/m3.

Table 3 detailed inorganic component emission factors

Table 3 reports PIXE (n=2) and IC (n=3) component emission factors as averages of steady and sooting burn experiments for each candle. Values are copied as µg h-1.

Source interpretation of metals

• The authors state that very low metal emissions were found for Candle I.
• For Candle II, the main metal compounds detected by PIXE were Cu, Sn, and Co. The authors suggest these may come from wick hardeners or wax additives such as color pigments, noting that Candle II was dark blue and that detected Cu and Co may have originated from color pigments.
• Wick/wax screening by PIXE found only Zn at very low levels in the Candle II wick; other compounds were below the detection limit, Sn was not analyzed in the wick measurements, and no compound was detected with PIXE in the wax.
• Size-resolved PIXE measurements found that for Candle II the only detected impactor components were potassium, tin, and copper; virtually all detected mass was below 1 µm.

Methods (brief)

Candle smoke was generated in a controlled 21.6 m3 stainless-steel chamber with filtered supply air. Four candles of one type were burned in each chamber experiment. Particle mass was measured by TEOM, particle size distributions by SMPS and volatility-SMPS, carbon fractions by thermal-optical OC/EC analysis, water-soluble ions by ion chromatography, elements by PIXE using stacked-filter and impactor samples, and morphology by TEM. Emission factors were calculated using a well-mixed box model with particle-loss correction. The detailed inorganic composition in Table 3 combines PIXE and IC results and is reported as component emission factors, not as candle-product concentration.

Related evidence

• Other home-air emission products (case-by-case, inhalation prioritized)
• Copper
• Tin
• Cobalt
• Zinc
• Lead

Update history

The five most recent substantive edits to this page, classified major (evidence or structure moved), correction (a published value or statement was wrong and has been fixed), or minor (narrative rewritten without changing the underlying evidence). Each description is derived from what the edit did to this page; the linked commit is the authoritative record, routine regeneration passes are excluded, and the full version history lives in git.